CN104202813A

CN104202813A - Cognitive radio network clock synchronization method based on double control channel mechanism

Info

Publication number: CN104202813A
Application number: CN201410462726.7A
Authority: CN
Inventors: 韩方剑; 余莉; 吴克宇
Original assignee: National University of Defense Technology
Current assignee: National University of Defense Technology
Priority date: 2014-09-12
Filing date: 2014-09-12
Publication date: 2014-12-10
Anticipated expiration: 2034-09-12
Also published as: CN104202813B

Abstract

The invention belongs to the field of wireless communication signal processing and discloses a cognitive radio network clock synchronization method based on a double control channel mechanism. The method includes firstly, performing the neighbor / level discovery and clock synchronous initialization based on the low-level control channel; secondly, performing the rough clock synchronization based on the low-level control channel; thirdly, performing the fine clock synchronization based on the high-level control channel. According to the method, the control channel establishment of a radio network is combined with the clock synchronization organically, the time information of the process of clock synchronization can be obtained owing to the establishing process of the control channels, the control channel establishment can be accelerated, the usability of the radio network in the segment of network establishment can be improved, and the application of the cognitive radio network can be promoted.

Description

Cognitive radio network clock synchronization method based on double-layer control channel mechanism

Technical Field

The invention belongs to the field of wireless communication signal processing, and particularly relates to a clock synchronization method for a cognitive radio network.

Background

A Cognitive Radio Network (CRN) is a novel wireless Network communication system, and is composed of Cognitive Radio devices called Cognitive Users (CUs) connected by wireless. The cognitive radio network is very different from a conventional wireless communication network in operation mode, and can coexist with a conventional wireless communication device called a Primary User (PU) and a system thereof in the same radio frequency band of the same area without causing harmful influence on the latter. There is no need to apply for operating band grants (or only a very small portion of the band grants) to the local radio spectrum management department before the cognitive radio network is established and operated. This characteristic of cognitive radio networks has attracted extensive, continuous and close attention from the academic world, the industrial world, commercial and military users, and the radio management departments of various countries in the field of wireless communications, and related technologies have developed rapidly in recent years.

Technically, the following steps are roughly required for safely establishing and effectively operating the cognitive radio network: firstly, cognitive users in a network collect and analyze local wireless spectrum use state information and detect peripheral conventional wireless equipment by means of a spectrum sensor, a network database and the like; then, making a decision according to the communication requirement of the cognitive radio network, and making a subsequent radio behavior scheme; and finally, by applying flexible wireless waveforms and communication protocols, all cognitive users in the cognitive radio network implement the radio behavior scheme in a coordinated and consistent manner to complete network communication tasks.

The implementation of the above technical approach needs to solve a plurality of key technical problems, including wireless spectrum monitoring, parameter estimation of wireless channel, detection and identification of adjacent cognitive users, and network clock synchronization. Since the cognitive radio network does not have the usage authorization of the fixed frequency band (or only has the usage authorization of a few partial frequency points), it has great technical difficulty in solving the problems. Particularly, in the initial establishment stage of the cognitive radio network, on one hand, since the cognitive user lacks key information such as available frequency points, network topology, network time reference and the like, and on the other hand, in order to avoid interference to the master user, the cognitive user cannot use communication auxiliary means such as conventional channel detection, handshake signaling and the like, the challenges of the technical problems are particularly prominent. The academics refer to these problems as "cognitive radio network establishment problems" (hereinafter referred to as network establishment problems) and have conducted extensive research.

In the early days, we proposed a network establishment method based on a dual-layer control channel mechanism for the network establishment problem and obtained a national invention patent (patent number ZL 201110259192.4). The method has the main technical thought that the control Channel is divided into a Lower Level Control Channel (LLCC) with limited power spectral density and a High Level Control Channel (HLCC) with high Channel capacity, and different wireless waveforms and communication protocols are adopted in the two control channels, so that the technical difficulty of the network establishment problem can be effectively reduced. In a low-layer control channel, the adoption of a waveform design with interference temperature limitation can ensure that a cognitive user and a master user coexist in the same region and the same frequency band, so that a cognitive radio network can be established and continuously operated under the condition of lacking spectrum authorization. On the basis that the basic control information exchange is realized by the low-layer control channel, the high-layer control channel is further established, the coverage area of the control channel is expanded, and the exchange rate of the control information is improved, so that the effective establishment of the cognitive radio network is realized step by step.

In the research and application process of the cognitive radio network based on the method, the network clock synchronization technology is found to be one of the necessary key technologies for solving the problem of establishing the cognitive radio network. The reason is that firstly, the operation of the cognitive radio network depends on spectrum sensing, spectrum use information is acquired by spectrum sensors distributed in various cognitive users, and a network-wide synchronous clock is necessary for effectively fusing the spectrum information to form a free spectrum space-time distribution graph available for the whole network. Secondly, in order to more efficiently utilize the detected idle frequency band, establish a high-speed cognitive channel and improve the throughput rate of the whole network, cognitive users must have a time division multiplexing/duplex mechanism for the cognitive channel; meanwhile, as the cognitive channels are randomly distributed in time and space, the accuracy and the stability of the time division multiplexing/duplexing mechanism can be ensured only by establishing the synchronization of the whole network clock. In addition, the clock synchronization of the whole network is undoubtedly beneficial to extracting key waveform parameters such as symbol synchronization, bit synchronization, frame synchronization and the like among cognitive users, and the point-to-point transmission performance can be improved.

However, if the clock synchronization process and the cognitive radio network establishment process based on the dual-layer control channel mechanism are separately and independently designed, the fact that the clock synchronization process and the cognitive radio network establishment process are mutually dependent and mutually constrained can be found. In one aspect, the distribution and exchange of clock synchronization information is dependent on the establishment of a higher layer control channel. Because the waveform power of the low-layer control channel is limited, the coverage area is small, the channel capacity is low, and a plurality of cognitive users share the low-layer control channel in a time division manner, the low-layer control channel can only be used for realizing some services with low requirements on data rate, such as neighbor/hierarchy detection (neighbor/hierarchy discovery). For the whole network clock synchronization process, if only a low-level control channel is used, the synchronization precision and the refresh rate are necessarily greatly limited, so that the support of a high-speed channel is required. On the other hand, the establishment of a high-speed channel depends on clock synchronization information with a certain accuracy. For example, adjacent cognitive users need to negotiate a commonly available idle frequency point, or called a spectrum hole (spectrum hole), in real time; high data rate communication waveforms and protocols require higher accuracy for symbol synchronization, bit synchronization, frame synchronization clocks, and the like.

In view of the above, we further propose a cognitive radio network clock synchronization method based on a dual-layer control channel mechanism. The main technical route of the method is that LLCC is used for providing low-speed control information exchange capability, the whole network clock synchronization initialization process is completed while the network initialization processes such as neighbor/level detection and the like are realized, and low-precision clock synchronization is provided to meet the requirement of establishing a high-level control channel. And then, realizing high-precision whole-network clock synchronization in the establishment process of the high-level control channel. Compared with the traditional wireless network clock synchronization method, the method can provide and maintain the whole network clock synchronization in the establishment process of the double-layer control channel, fully utilizes the channel resources of the double-layer control channel, and gradually improves the clock synchronization precision. The characteristic is very helpful to ensure the coordination of the wireless behaviors of the cognitive users in the cognitive radio network, improve the efficiency of network establishment and improve the stability of network operation.

Disclosure of Invention

The invention discloses a cognitive radio network clock synchronization method based on a double-layer control channel mechanism, aiming at the problem of network clock synchronization in the development and application of a cognitive radio network, and the method comprises the following steps:

is provided withThe cluster head local time of the nth cluster head broadcast message is represented, and n is a natural number; setting a given user cluster in a cognitive radio network to contain at least one cognitive user;

firstly, neighbor/layer discovery and clock synchronization initialization based on a low-layer control channel;

step (S11), a certain cognitive user is set to wait for the cluster head broadcast message in a low-level control channel, and if the cluster head broadcast message is not detected, the cognitive user is determined to be the cluster head;

step (S12), the cluster head sends the broadcast message of cluster head through the low-layer control channel, the broadcast message of cluster head includes the ID number and local sending time of the cognitive user

Step (S13), the cognitive users except the cluster head read the time from the broadcast message of the cluster headAnd recording the arrival time of the cluster head broadcast messageSimultaneously returning an adding request message or an exiting request message to the cluster head, wherein the adding request message or the exiting request message comprises the ID number and the local sending time of the cognitive user

Step (S14), the cluster head receives the join request message or the quit request message from a cognitive user through the low-level control channel, and records the arrival time of the messageAt the moment, the cognitive user is recorded as a cluster member; then, a reply including the local sending time is sent to the cluster member sending the messageAllow join message or allow exit message;

in the step (S15), after receiving the join allowing message or the quit allowing message returned by the cluster head in the step (S14), the cluster members record the local arrival timeAnd reading the timeAndcalculating to obtain an estimated initial value of cluster head clock parameters and cluster broadcast message propagation delay

Step (S16), the cluster members obtain the estimated value of the cluster head clock through calculation, and the clock synchronization initialization between the cluster members and the cluster heads is completed;

secondly, clock coarse synchronization based on a low-layer control channel;

step (S21), the cluster head periodically transmits N-1 cluster head broadcast messages through the low-layer control channel, and the transmitted messages include local transmission time dataWherein N is { 2., N }, and N is a natural number greater than or equal to 2;

step (S22), the cluster member receives the cluster head broadcast message and records the local receiving time asAnd reading the time

Step (S23), updating the clock parameter estimation value of the cluster member to the cluster head by combining the initial value of the cluster member to the local clock parameter estimation of the cluster head and the initial value of the cluster broadcast message propagation delay obtained in the step one;

step (S24), updating the estimated value of the cluster head local clock of the cluster members;

thirdly, the clock is precisely synchronized based on a high-level control channel;

two cluster members exist in a certain cluster of the cognitive radio network and are respectively marked as CM_AAnd CM_BAnd they have already finished the clock synchronization initialization of the said first step and clock coarse synchronization of the second step with the cluster head separately;

step (S31), the cluster head continues to send the broadcast message of cluster head on the low-level control channel, and writes the local time in the messageWherein N belongs to { N + 1.,. N + m }, and m is a natural number;

step (S32), CM_AAnd CM_BReceiving cluster head broadcast message on low-layer control channel, readingAnd respectively record the receiving time asAnd

step (S33), CM_BUpward CM on higher layer control channel_ASending a local time writeAnd cluster head timeSending the request message;

step (S34), CM_AReceiving CM over higher layer control channel_BSending request message of recording arrival timeAnd reads out time data in the transmission request messageAndaccording toCM_AFinding corresponding cluster head broadcast message arrival time from local database

Step (S35), CM_ATo CM_BSending a write withAnd local transmission timeThe transmission permission message of (1);

step (S36), CM_BReceiving the transmission permission message, recording the local arrival timeAnd read outAnd

step (S37), calculating CM_BAnd CM_AAn estimate of a relative clock parameter therebetween;

step (S38), calculating CM_BTo CM_AEstimate of local clock, complete CM_AAnd CM_BBased on the clock fine synchronization process of the high-level control channel.

Further, in the first step (S15), an initial value of the cluster member local clock parameter estimation for the cluster head is calculatedInitial value of estimation of propagation delay of sum cluster broadcast messageThe method specifically comprises the following steps:

wherein,

further, the specific calculation process in the second step (S23) is:

wherein,

A^{(n)} = [\begin{matrix} T_{1}^{(1)} + {\hat{D}}^{(1)} \\ T_{1}^{(2)} + {\hat{D}}^{(1)} \\ . \\ . \\ . \\ T_{1}^{(n)} + {\hat{D}}^{(1)} \end{matrix}], B^{(n)} = [\begin{matrix} T_{2}^{(1)}, & - 1 \\ T_{2}^{(2)}, & - 1 \\ . & . \\ . & . \\ . & . \\ T_{2}^{(n)}, & - 1 \end{matrix}],

represents the nth-order estimate of Θ, N ═ 1, ·, N },is an estimate of the clock parameter.

Further, the specific calculation process in the second step (S24) is:wherein,n-th estimated value, t, representing local clock of cluster head to cluster member_CMIs the clock value of a member of the cluster,is an estimate of the clock parameter.

Further, the calculation formula of the (S37) th step in the third step is:

wherein,

representation CM_AAnd CM_BAfter receiving the N + m times of cluster head broadcast message, CM_AAnd CM_BThe mth relative clock parameter estimate in between.

Further, the calculation formula of the (S38) th step in the third step is: CM (compact message processor)_BEstimating CM_ALocal clock of

Wherein,is CM_BAnd CM_AThe mth relative clock parameter estimate in between.

The basic idea of the invention is as follows: the whole network clock synchronization process of the cognitive radio network is divided into three steps, which are respectively as follows: neighbor/level discovery and clock synchronization initialization based on LLCC, clock coarse synchronization based on LLCC and clock fine synchronization based on HLCC. The specific principle and the derivation process of the related calculation formula are as follows:

it is assumed that the non-idealities of the cognitive user's local Clock are modeled as Clock Skew (Clock Skew) and Clock Offset (Clock Offset). I.e. any two cognitive users CR_iAnd CR_jLocal clock t of_iAnd t_jSatisfy t_i＝f_ijt_j+τ_ijWherein f is_ijIs CR_iRelative to CR_jClock tilt of τ_ijIs CR_iRelative to CR_jClock skew of (2). And exchanging signaling messages written with local time stamps among the cognitive users. Depending on the environment in which the cognitive radio network is located, the time delay variable for a signaling message to propagate from one user to another can be modeled as one of two random processes: i.e. a gaussian distribution random process and an exponential distribution random process for a given mean. Since the modeling manner of the time delay variable does not affect the operation method of the cognitive user in the network establishment and clock synchronization processes, but only affects the final calculation formula of the clock parameter, herein, for convenience of describing the detailed process of the present invention, the following derivation is performed by taking a gaussian distribution random process as an example.

First, neighbor/level discovery and clock synchronization initialization process based on LLCC.

For any given user cluster in a cognitive radio network that contains at least one cognitive user, the neighbor/stratum discovery and clock synchronization initialization phases operate as follows. The LLCC adopts a time division multiplexing manner, and includes three kinds of time slots, which are a Cluster Head Broadcast (CHB) time slot, a Cluster Member Broadcast (CMB) time slot, and a Cluster head response (CHA) time slot. In the CHB time slot, the cluster head CH broadcasts and writes local time on the LLCCOther cognitive users CU listen to the message at the time slot and record the CU local time of the received cluster head broadcast messageAnd the CH local time for transmitting the broadcast message from the cluster head read from the messageFor CUs that want to join or exit a cluster, the written CU local time can be broadcast on the LLCC at the CMB slotA Join Request (Request To Join, RTJ) message or an exit Request (Request To Quit, RTQ) message. When receiving RTJ or RTQ message, CH firstly records message arrival timeThen, a local time written with CH is recovered in the CHA time slotAnd the time of arrival just recordedIs allowed To Join (Clear To Join, CTJ) messages or is allowed To exit (Clear To Quit, CTQ) messages. The CU receives CTJ or CTQ to complete the hierarchical discovery of the cognitive user in the cluster, and simultaneously needs to complete the hierarchical discoveryRecording message arrival timeAnd reading the time in the message sent by the cluster headAndto complete the initialization of clock synchronization with the CH. Defining the clock tilt and clock offset of cluster head CH and certain cluster member CM as f and tau respectively, then cluster head clock t_CHClock t with cluster member_CMThe following relationship is satisfied:

t_CH＝(t_CM- τ)/f (equation 1)

And time data recorded by CMSatisfies the following relationship, wherein D⁽¹⁾Is the message propagation delay.

(formula 2)

(equation 2) can be rewritten as:

(formula 3)

Wherein theta is₁＝1/f，θ₂＝τ/f。θ₁，θ₂Which is indicative of a parameter of the clock,denotes theta₁The estimated initial value of (a) is,denotes theta₂The estimated initial value of (equation 3) is rewritten into a matrix form

Α⁽¹⁾＝B⁽¹⁾X (formula 4)

Wherein,

thus, the clock parameters can be estimated as follows:

(formula 5)

Thus, the CM obtains an estimated initial value of the CH local clock

Secondly, clock coarse synchronization based on LLCC:

after the layer discovery of the cluster in the cognitive network is completed and the clock initialization is carried out, a cluster member CM obtains a cluster head CH clock parameter theta₁，θ₂And propagation delay D⁽¹⁾Estimated initial value ofAndmeanwhile, the CH continues to transmit the CH broadcast message through the LLCC. Assuming that the CH continues to send the CH broadcast message for N-1 times after the CH completes the first step operation, wherein N is a natural number with the value more than or equal to 2; if the CH broadcast message in the first step is also calculated, the CH local time data in all CHB broadcast messages can be recorded asn∈[1,N]. The CM receives the broadcast messages of the CH and records the arrival time of each broadcast message asn∈[1,N]. The relationship of the CH clock and the CM clock may be expressed as

(formula 6)

Wherein,is a zero mean random variable of a gaussian distribution. (equation 6) can be written in the form of a matrix as follows.

Α⁽ⁿ⁾＝B⁽ⁿ⁾Θ+d⁽ⁿ⁾(formula 7)

Wherein,

can further obtain:

(formula 8)

The nth estimated value representing Θ gradually improves the estimation accuracy as n changes,is the nth estimate of the CH clock parameter by CM.

Thus, the CM gets an nth estimate of the CH local clock:wherein, t_CMIs the local time of the cluster member.

Thirdly, clock fine synchronization based on HLCC:

if there are two cluster members (denoted as CM) in a cluster of a cognitive radio network_AAnd CM_B) Then CM_AAnd CM_BThe clock synchronization initialization and clock coarse synchronization processes for the CH have been completed separately. Without loss of generality, set CM_AAnd CM_BN CHB messages are received. If HLCCs need to be established between them to support subsequent communication tasks, they can establish the information for CMs in currently available spectrum holes using a control channel convergence algorithm with time synchronization conditions (such algorithm can adopt the effective scheme in the prior art, and is not in the scope of the discussion herein)_AAnd CM_BAvailable HLCC. In HLCC establishment process, CM_AAnd CM_BCan accomplish mutual clock fine synchronization process between, specific process is:

the CH continues to periodically send to the entire cluster via LLCC with local send timeIn this case, N is ∈ { N +1, N +2,. and N + m }, where m is a natural number. CM (compact message processor)_AReceiving CHB message, readingData and recording local reception timeCM_BReceiving the broadcast message, reading the transmission timeAnd recording the local time of receptionAt the same time, CM_BAnd CM_APeriodically trying to exchange control signaling messages (the same period as the CHB broadcast period). With CM_BTo CM_ASending a local time writtenAnd corresponding CHB message transmission timeRequest To Send (RTS) message. If CM_AWhen the RTS message is received, the local time of arrival of the message is recordedAnd reading out time data in the messageAndaccording toCM_AFind the arrival time corresponding to the CHB message sending time from the local databaseCM_ATo CM_BSending a local time writtenAndandclear To Send (CTS) message. If CM_BUpon receiving the CTS message, the time of arrival is recordedAnd reading out time dataAnd

for a certain time N ∈ { N +1, N +2, ·, N + m }, CM_B、CM_AThe relationship between the local clocks of the three and the CH is as follows.

(formula 9)

Wherein,f_Aindicating cluster head CH and cluster member CM_AClock tilt of f_BIndicating cluster head CH and cluster member CM_BClock tilt of t_ARepresenting cluster members CM_AClock of (t)_BRepresenting cluster members CM_BClock of τ_AIndicating cluster head CH and cluster member CM_AClock offset of τ_BIndicating cluster head CH and cluster member CM_BClock offset of t_CHIndicating a cluster head clock;

the two formulas above are combined as:

t_A＝t_Bφ₁+φ₂(formula 10)

Wherein, theta_A、θ_BRespectively representing cluster members CM_AAnd cluster member CM_BCorresponding clock parameters; let theta_A＝1/f_A，θ_B＝1/f_B；

Thus, phi₁、φ₂Referred to as CM_AAnd CM_BRelative clock parameters therebetween; further, the relationship of obtaining the corresponding time data is as follows:

(formula 11)

(formula 12)

(formula 13)

(formula 14)

WhereinAndrespectively representing slave CM_ATo CM_BAnd a slave CM_BTo CM_AThe propagation delay of (2). Can getAdding (equation 13) and (equation 14) to obtain

(formula 15)

The difference between (equation 11) and (equation 12) is added to (equation 15) to obtain

(formula 16)

Note that (equation 11) and (equation 12) also hold for all N ∈ {1, 2.,. N } time instants, so subtracting them yields:

<math> <mrow> <msubsup> <mi>T</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>A</mi> </mrow> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <msubsup> <mi>T</mi> <mrow> <mn>2</mn> <mo>,</mo> <mi>B</mi> </mrow> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msubsup> <mo>·</mo> <msub> <mi>φ</mi> <mn>1</mn> </msub> <mo>+</mo> <msub> <mi>φ</mi> <mn>2</mn> </msub> <mo>+</mo> <mrow> <mo>(</mo> <msubsup> <mi>D</mi> <mi>A</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msubsup> <mo>-</mo> <msubsup> <mi>D</mi> <mi>B</mi> <mrow> <mo>(</mo> <mi>n</mi> <mo>)</mo> </mrow> </msubsup> <mo>)</mo> </mrow> <msub> <mi>f</mi> <mi>A</mi> </msub> <mo>,</mo> <mi>n</mi> <mo>&Element;</mo> <mo>{</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <mi>N</mi> <mo>+</mo> <mi>m</mi> <mo>}</mo> </mrow> </math>

(formula 17)

Writing (equation 16) and (equation 17) in a matrix form

P^(m)＝Q^(m)Φ^(m)+ξ^(N+m)(formula 18)

Wherein

It is possible to obtain:

(formula 19)

To this end, CM_BFinish to CM_AEstimation of local relative clock parameters. In the same way, CM_ACan also obtain the pair CM_BEstimation of local relative clock parameters.

CM can be obtained according to m-th phi parameter estimated value_BTo CM_AThe mth estimation of the local clock is

This is CM_AAnd CM_BThe clock fine synchronization process based on the HLCC is carried out between the clock fine synchronization process and the clock fine synchronization process.

The beneficial effects obtained by adopting the invention are as follows:

the invention organically combines the establishment of the control channel of the cognitive radio network with the clock synchronization. The whole network synchronous clock information provided by the time synchronization process can be continuously obtained in the establishment process of the control channel, so the establishment process of the control channel can be accelerated; the time synchronization process of the invention can exchange the local time information more efficiently after being supported by the high-speed control channel, thereby continuously improving the precision of time synchronization. The invention provides an effective cognitive radio network time synchronization method, which improves the usability of a cognitive radio network in a network establishment stage and is beneficial to the practical application of the cognitive radio network.

Drawings

FIG. 1 is a flow chart of a cognitive radio network clock synchronization method based on a double-layer control channel mechanism;

FIG. 2 is a schematic diagram of message exchange of a first step and a second step of cognitive radio network clock synchronization based on a double-layer control channel mechanism;

fig. 3 is a message exchange diagram of the third step of cognitive radio network clock synchronization based on a double-layer control channel mechanism.

Detailed Description

The invention is further described with reference to the following figures and specific examples.

Fig. 1 is a flow chart of cognitive radio network clock synchronization based on a dual-layer control channel mechanism. The whole process is divided into three steps.

In the first step, the specific steps of the hierarchical discovery and clock initialization based on the LLCC are (n ═ 1).

(1) A certain cognitive user CU waits for broadcasting of a cluster head CH in the LLCC, and if the broadcasting of the CH is not detected, the cognitive user sends a CHB message (including the ID number and the local sending time of the CU) through the LLCC) Declares itself a cluster head CH.

(2) When a CU receives a broadcast from a CH, the CU sets a cognitive user sending a CHB message as the CH, recognizes the cognitive user as a cluster member CM of the CH, and reads CH local time data from the CHB messageAnd records the arrival time of the CHB messageSimultaneously returning an RTJ message to CH, and storing its own ID and local time dataWritten into the message.

(3) The CH receives RTJ message from a CU through LLCC and records the arrival time of the messageAnd reading the ID number data from the message, and setting the CU as a member of the cluster.

(4) The CH replies CTJ the message to the CM through the LLCC and sends its own local delivery timeAnd recorded RTJ message arrival timeWritten into the message.

(5) The CM receives CTJ from CH via LLCC and records message arrival timeAnd read the time in the messageAndtherefore, the CM obtains an estimated initial value of the CH local clock parameter and the CHB message propagation delay:

(6) the CM obtains an initial estimate of the CH local clock:

in the second step, the specific steps of coarse clock synchronization based on LLCC are described as follows (N ═ 2.., N }).

(1) The CH periodically transmits CHB messages with local time data written therein using LLCC

(2) The CM receives the CHB message and keeps the local receiving time asAnd reads the CH local time data in the message

(3) The CM estimates the clock parameter of the CH to obtainAndthe calculation formula is shown in (formula 8).

(4) The CM obtains an estimate of the CH local clock:

the third step, the detailed step of the clock fine synchronization based on the HLCC is as follows (N ═ N + 1.

(1) The CH continues to send CHB messages on the LLCC with local time data written therein

(2)CM_AAnd CM_BCHB message received on LLCC, readAnd respectively record the receiving time asAnd

(3)CM_Bupstream CM on HLCC_ASending a local time writeAnd CH timeThe RTS message of (1).

(4)CM_AReceiving RTS message on HLCC, recording arrival timeAnd reads time data in the RTS messageAndaccording toCM_AFinding corresponding CHB message arrival time from local database

(5)CM_ATo CM_BSending a local time writtenAndandthe CTS message of (2).

(6)CM_BReceiving the CTS message, recording the time of arrivalAnd read outAnd

(7)CM_Bget a pair of CM_ARelative clock parameter ofAndthe calculation formula is shown in (formula 19).

(8) Calculating CM_BTo CM_AThe local clock of is

Fig. 2 is a schematic diagram of message exchange of the first step and the second step of cognitive radio network clock synchronization based on a double-layer control channel mechanism. As shown, the local clocks of CH and CM are represented as time axes t_CHAnd t_CMThe clock offset of both is τ, and the clock tilt of CM with respect to CH is f. The first step, CHB, CMB and CHA messages are exchanged between CH and CM in turn, the corresponding message sending time is respectivelyAndthe message receiving times are respectivelyAndin the second step, the CH periodically sends CHB messages to the CM for a period of timen∈[2,N]At a receiving time ofn∈[2,N]。

Fig. 3 is a message exchange diagram of the third step of cognitive radio network clock synchronization based on a double-layer control channel mechanism. In the figure, CH and CM_AAnd CM_BRespectively, as a time axis t_CH、t_AAnd t_B。t_AAnd t_CHIs offset by tau_A，t_BAnd t_CHIs offset by tau_B. Clock skew of CMA with respect to CH is f_A，CM_BClock skew with respect to CH of f_B. The CH periodically transmits CHB messages with a transmission time set toCM_AAnd CM_BThe local time of receiving the CHB message is respectively set toAndCM_Aand CM_BAnd RTS and CTS signaling messages are exchanged therebetween in sequence. CM (compact message processor)_BWhere the local time to send the RTS message is set toLocal time of receiving CTS message is set toCM_AWhere the local time to receive the RTS message is set toLocal time to send CTS message is set toThe N is formed by { N +1, N +2,.., N + m }, and m is a natural number.

The above embodiments are only for illustrating the technical solutions of the present invention and are not limited, and it should be understood by those skilled in the art that modifications or equivalent substitutions can be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and the technical solutions of the present invention should be covered in the claims of the present invention.

Claims

1. A cognitive radio network clock synchronization method based on a double-layer control channel mechanism is characterized by comprising the following steps:

in the step (S15), the cluster members receive the join allowing message returned from the cluster head in the step (S14)After allowing the exit message, recording the local arrival timeAnd reading the timeAndcalculating to obtain an estimated initial value of cluster head clock parameters and cluster broadcast message propagation delay

secondly, clock coarse synchronization based on a low-layer control channel;

2. A substrate as claimed in claim 1The cognitive radio network clock synchronization method based on the double-layer control channel mechanism is characterized in that in the first step (S15), the initial value of the cluster member to the local clock parameter estimation of the cluster head is calculatedInitial value of estimation of propagation delay of sum cluster broadcast messageThe method specifically comprises the following steps:

wherein,

3. the method for synchronizing cognitive radio network clock based on the dual-layer control channel mechanism as claimed in claim 1, wherein the step (S23) in the second step comprises the following specific calculation processes:

wherein,

A^{(n)} = [\begin{matrix} T_{1}^{(1)} + {\hat{D}}^{(1)} \\ T_{1}^{(2)} + {\hat{D}}^{(1)} \\ . \\ . \\ . \\ T_{1}^{(n)} + {\hat{D}}^{(1)} \end{matrix}], B^{(n)} = [\begin{matrix} T_{2}^{(1)}, & - 1 \\ T_{2}^{(2)}, & - 1 \\ . & . \\ . & . \\ . & . \\ T_{2}^{(n)}, & - 1 \end{matrix}],

4. The method for synchronizing cognitive radio network clock based on the dual-layer control channel mechanism as claimed in claim 1, wherein the step (S24) in the second step comprises the following specific calculation processes:wherein,n-th estimated value, t, representing local clock of cluster head to cluster member_CMIs the clock value of a member of the cluster,is an estimate of the clock parameter.

5. The cognitive radio network clock synchronization method based on the dual-layer control channel mechanism as claimed in claim 1, wherein the third step (S37) is a step of calculating the formula:

wherein,

6. The cognitive radio network clock synchronization method based on the dual-layer control channel mechanism as claimed in claim 1, wherein the third step (S38) is a step of calculating the formula: CM (compact message processor)_BEstimating CM_ALocal clock of

Wherein,is CM_AAnd CM_BThe mth relative clock parameter estimate in between.